INDEX

Explanations

concepts and phrases that emphasize completeness and universality in datasets or models

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Réponses

-0.51

小小

-0.50

小小的

-0.49

 dokun

-0.48

pé

-0.48

ParseTree

-0.48

 semp

-0.47

 simpler

-0.46

<eos>

-0.46

få

-0.46

POSITIVE LOGITS

 AssemblyCulture

0.76

mybatisplus

0.73

 myſelf

0.68

 indisponible

0.68

 unrestricted

0.65

 fhew

0.64

 itſelf

0.62

็จ

0.62

NUMX

0.61

 defire

0.60

Activations Density 0.549%

concepts and phrases that emphasize completeness and universality in datasets or models

No Comments

No Known Activations